(895) Immune Phenotyping Results in Pediatric Sepsis Patients: Impact of Preprocessing Time Intervals

Introduction: Sepsis is characterized by phenotypes with diverse pathobiology. Emergency department biospecimens reflect acute sepsis biology (often prior to treatment), but the logistics of sample processing across centers is complex. In preparation for a multicenter study, we completed a single-center pilot to determine the maximum pre-processing interval for sepsis blood samples which would yield accurate results. We hypothesized that cellular immune phenotype and inflammatory proteins would be altered by increasing time intervals prior to processing.

Methods: In 7 children with sepsis, we collected a blood sample within 3 hours of sepsis recognition. Blood was divided into 3 aliquots for processing: within 3 hours (reference), within 12-24 hours (to simulate overnight collection), and within 48-72 hours (to simulate shipping and processing at a central site). We isolated peripheral blood mononuclear cells (PBMC) to analyze immune profiles via a 35-marker spectral flow cytometry panel and 4 cellular metabolic states by Single Cell ENergetIc metabolism by profiling Translation inHibition (SCENITH). We measured expression of 365 inflammatory proteins by Olink proximity extension assay. Results were compared across time-dependent processing conditions by paired t test, ANOVA, and ordinal regression.

Results: The composition and activation of T and B cell subsets and cellular metabolic states did not vary by pre-processing interval. At 12-24 hours, PBMC immune profiles were not significantly different compared to 3-hour pre-processing time, but after 48-72 hours, the proportion of HLA-DR+ monocytes increased while the quantity of some innate inflammatory cells (NKT, MDSC) decreased (all p< 0.05). Longer pre-processing time interval was associated with increased expression of pro-inflammatory mediators, including MPIG6B, SKAP2, TBC1D5, CXCL8, MANF, OSM (all p< 0.05 compared to 3 hours).

Conclusions: Longer time interval prior to blood processing was associated with increased monocyte HLA-DR expression, loss of some innate inflammatory cells, and amplified pro-inflammatory protein expression. While PBMCs were minimally affected by a simulated “overnight” processing delay, proteomic analysis is best performed on fresh samples. These results inform multicenter study approaches to sepsis immune phenotyping.